INSTRUCTIONS FOR RUNNING CODE

1. run the "Cleaning Code" R code. You need to save the raw data (the .gz) and the .xml to your working directory 
	before running it. This is the raw data that everything is run off of + the "instructions" for R to read 
	the raw data. You do not need to unzip the .gz file.
2. run the GOV2001_Checkpoint3..." R code. This contains our replication of the paper AND 
	our own analyses/extensions.
3. The "simulating risk differences" R code is what we use to generate risk differences/DID estimates
	in the original scale (e.g., to avoid IRRs/ORs for Poisson/binomial models). It also creates 
	the figures showing risk differences for the various subgroups. Because it is long, we put it in 
	a separate script.

Reach out with questions!
 